The identification of disease susceptibility genes in human families and populations to date has focused on the mapping and cloning of autosomal dominant highly penetrant genes, identified by linkage mapping studies using extreme, "high risk families." Such families are rare and because of their small size often provide limited genetic information, particularly if key family members are deceased or have not produced offspring. In addition, inherent limitations in the structure of human populations, such as a long generation time and outbreeding mean that genes which are weakly penetrant, give rise to variable phenotypes, or act in concert with other genes to produce complex phenotypes are extremely difficult to map. This is particularly true for diseases like cancer, in which a multitude of low penetrant alleles are believed to account for a significant percentage of disease in the population.We hypothesized previously that the domestic dog offers several key advantages over other systems for mapping genes relevant to human disease. Obvious advantages include the large family size of dogs, as well as founder effects and narrow population bottlenecks that result in limited locus heterogeneity for common disorders among purebred dogs.In this grant, three experienced and interactive groups will expand their ongoing collaborations to develop the infrastructure for positional cloning in dogs, thus establishing the domestic dog as the model system of choice for mapping cancer susceptibility genes, as well as other common disease alleles. Previously we developed an integrated 1800 marker meiotic linkage/radiation hybrid (RH) map of the dog, composed of microsatellites, ESTs, and BAC-ends that spans over 95 percent of the genome. We have demonstrated the utility of these maps by mapping loci for several diseases, including canine renal cancer and four types of retinal degeneration. In this proposal we will set the stage for positional cloning of these and other disease genes of interest in the dog by 1) Identifying fragments from greater than 10,000 canine genes, suitable for RH mapping, from the 1x canine genome sequence being made available by The Institute for Genomic Research (TIGR); 2) Mapping fragments from 10,000 canine genes using a newly developed, high resolution (10,000-rad) RH panel; and 3) Developing and making readily available to the genomics community a canine RH map inclusive of these and additional data currently under generation.